Savings & Evaluation Methodology for Codes and...
Transcript of Savings & Evaluation Methodology for Codes and...
Savings & Evaluation Methodology for Codes and Standards Initiative
Submitted to:
Massachusetts Department of Energy Resources
On behalf of:
Massachusetts Program Administrators
Commercial/Industrial & Residential Energy Efficiency Programs
October 20, 2015
Mass Save ®is a statewide initiative sponsored by:
Berkshire Gas |Cape Light Compact| Columbia Gas of Massachusetts |Eversource|
National Grid|Liberty| Unitil
| MassSave.com |
Page 1 of 2
TABLE OF CONTENTS
Summary ..................................................................................................................................................... 2
1.0 Introduction & Background .................................................................................................................. 4
2.0 Savings Methodology for Code Compliance Support Initiative ........................................................... 7
2.1 Estimates of Gross Technical Potential ........................................................................ 10 (a) 2.1.1 Residential Sector .......................................................................................................... 10
(b) 2.1.2 Commercial Sector ........................................................................................................ 14
2.2 Estimates of Gross Savings .......................................................................................... 17 2.3 Attribution to PA Efforts and Net Savings ............................................................................. 18 3.0 Evaluation ........................................................................................................................................... 25
3.1 Previous Evaluation Efforts .................................................................................................... 25 3.2 Current and On-going Evaluation Efforts ............................................................................... 26 3.3 Future Evaluation Planning ..................................................................................................... 27
4.0 Appliance Standards Advocacy .......................................................................................................... 30
APPENDIX ........................................................................................................................................................... 32
Page 2 of 3
SUMMARY The Massachusetts Program Administrators (PAs) of the Commercial/Industrial & Residential Energy Efficiency
Programs are pleased to present this document titled “Savings and Evaluation Methodology for a statewide
Building Energy Codes and Appliance Standards (C&S) Initiative”. In April 2014, the Massachusetts PAs began
implementation for a large component of the PAs’ overall Codes and Standards Strategy: Code Compliance
Support Initiative (CCSI); and may begin advocacy work for appliance standards in 2016.
Claiming savings and attribution for codes and standards is a relatively new concept for PA’s across the country.
As a result, there are not a lot of historical methodologies to draw upon when developing a savings
methodology and attribution formula. This document specifically outlines a methodology that the PAs have
developed to calculate gross savings and attribution to the PA effort (also referenced as net energy savings), as
well as an evaluation plan related to the code compliance effort and appliance standards advocacy.
Due to the complexities detailed within this document regarding energy code non‐compliance in the retrofit
market, this savings methodology only addresses new construction at this time. The PAs will continue to work
with the EEAC Consultants and other national experts to determine how retrofit savings can be applied to this
methodology as there may be a higher non‐compliance rate in retrofits than new construction.
The need to evaluate the efforts put forward by the PAs is essential in determining the effect CCSI has on code
compliance savings and the future continued funding for the initiative. The PAs evaluation approach
incorporates baseline studies, surveys, data/material collection, and Delphi panels. Our efforts in evaluation will
consider other alternatives (besides PNNL approach for baseline studies or a modified version of this approach)
should they be deemed to result in a more accurate estimation of the impacts of the CCSI. Evaluation studies
will typically be conducted at the beginning and end of a given code cycle to best gauge the increase in code
compliance through the various determined components. Delphi panels will likely be used in the future and will
be responsible for analyzing available data and come to a consensus on savings through the CCSI and appliance
standards efforts.
While implementing the CCSI has a reasonable cost associated with it, evaluation costs tend to be much more
expensive due to the fact that they are time and labor intensive. Therefore while the PAs are committed to
collecting and analyzing data to ensure that savings calculations are as accurate as possible and the initiative
implementation builds on its experiences and successes, ensuring that costs are controlled is also a key
consideration.
The main purpose of this document is to initiate a discussion between Program Administrators and DOER so that
a collectively agreed upon path to evaluate the C&S effort and a savings methodology can be determined for the
2016‐2018 three year plan. This document was compiled based on several discussions with EEAC consultants
and national experts in the field of codes and standards implementation, evaluation, and advocacy.
Page 3 of 4
Page 4 of 5
1.0 Introduction & Background The Codes and Standards Initiatives are gaining momentum throughout the country. For example, California has
a long‐standing program in place that has resulted in significant energy savings from code development and
advocacy. Arizona utilities have recently begun a code compliance enhancement initiative, while other states1
are beginning to investigate and develop Codes and Standards activities.
Energy codes affect a building’s efficiency level at initial construction, during renovations and alterations and
when equipment is replaced. All buildings, whether residential, commercial, or industrial have a profound
impact on their surroundings, both environmentally and economically. The successful implementation of energy
codes leads to benefits that last for many decades. It is a process where the design, construction and
enforcement communities are equal players. Successful implementation is not always easy and requires
support. Without support, codes are not always implemented properly. For example, overall code compliance
for commercial new construction in Massachusetts (2012 evaluation study, also referenced below) was
estimated to have a 80% compliance rate, with none of the surveyed buildings fully in compliance with the
prevailing code.
Similarly, appliance standards work to ensure that the many appliances used every day are built to an efficiency
level that will benefit consumers, the environment, and the economy. Effective standards save energy, save
money for ratepayers through reduced energy and demand, reduce power plant emissions, and can create jobs
through higher demand for energy efficient products. Standards may be set at the federal or state level. The
PAs can also advocate for appliance standards at the national level.
The Massachusetts Clean Energy and Climate Plan, developed by the Executive Office of Energy and
Environmental Affairs to meet the requirements of the Global Warming Solutions Act, specifically cites
“Advanced Building Energy Codes” and “Appliance and Product Standards” as two areas that are being relied
upon to deliver energy savings and related greenhouse gas emissions reductions to meet the Act’s goals.
Based on the 2012 code baseline evaluation studies and all the reasons mentioned above, in 2012, the PAs
proposed a codes and standards implementation plan to DOER for the 2013‐2015 Three Year Energy Efficiency
Plan. At that time, the PAs did not have enough information or a robust enough methodology to support PA
savings through these implementation activities. As a result, starting 2014, the PAs started implementing the
‘code compliance support’ piece of the overall codes and standards initiative and did not claim any savings for
this effort. Given below are a list of activities the PAs have been implementing since 2014 and plan to continue
in 2016‐2018 program cycle.
1 Other known states currently investigating Codes and Standards include Connecticut, Vermont, New York, Maryland, Washington,
Oregon, Ohio, Minnesota, Colorado, and Illinois.
Page 5 of 6
1.1Scope
The scope of this memo is as follows:
Provide a very brief overview of implementation activities associated with CCSI.
Describe in detail the savings methodology for gross and net savings that can be attributed to PAs for
implementing code activities. Includes assumptions and modeling that went behind the savings
calculations, along with tables and appendix to offer more insights into the calculations.
Describe evaluation activities.
Provide a brief overview of appliance standards. Since this is not the main focus of this document, an
overview has been provided with more details to follow during the course of the 2016‐2018 program
year period.
1.2ImplementationActivitiesThe PAs are proposing two distinct initiatives in the 2016‐2018 program cycle under the C&S umbrella and they
are briefly described below. The PAs compliance support activities will continue to evolve during the course of
the initiative, based on implementation experience and evaluation results.
1. Code Compliance Support Initiative (CCSI):
Objective: The CCSI will increase the ability and desire of both the design community (architects and
engineers) and the construction community (builders, contractors, and construction managers), to
comply with the locally mandated building energy code; generally either the version of the IECC
adopted statewide or the stretch code. In addition, the CCSI will increase the ability and desire of
code officials and others to understand, monitor, document and ensure code compliance in their
jurisdictions. The scope of PA efforts will support the codes, generally including both new
construction and retrofits/renovations in existing buildings, and both residential and commercial
sectors. Specific activities include:
1. Trainings: The PAs will continue to utilize third party vendors to develop and deliver residential
and commercial classroom style trainings that are geographically dispersed around the state. Web‐
based residential and commercial focused trainings will also occur.
2. Energy Code Circuit Rider Technical Support: The PAs propose to continue utilizing Circuit Riders
as the providers of technical assistance for Massachusetts. The roles of Circuit Riders are: to act as
consultants (either via phone or in‐person) on energy codes and energy efficient building design and
practice while clarifying any confusion or misunderstanding that building design and construction
professionals may have about energy codes.
3. Documentation Tools: The PAs will develop and support consistent documentation tools such as
residential and commercial FAQ lists, new construction field guides, software tools, and checklists
Page 6 of 7
for adoption by jurisdictions as a means of compliance enhancement. A more consistent
expectation of documentation and formatting requirements will allow designers and builders to
more effectively communicate with plan reviewers and inspectors on how specific projects can
comply with the energy code.
2. Appliance Standards Advocacy Objective: The Appliance Standards Advocacy Initiative will accelerate the development and
adoption of selected new state appliance standards. The adoption of more stringent appliance
standards will help the state achieve higher energy efficiency savings while yielding additional
environmental benefits. Targeted residential and commercial appliances are excellent efficiency
candidates because often there are large numbers of each specific appliance sold in the state.
Specific activities include:
Collaborate with key stakeholders such as Northeast Energy Efficiency Partnership (NEEP),
Appliance Standards Assistance Project (ASAP), and other entities to determine a model
package of potential target appliances for state level adoption. A model package represents
a grouping of selected appliances which have the potential to yield substantial savings.
Work with stakeholders to collect key appliance data, conduct market potential for key
appliances and determine associated savings potential and societal costs and benefits.
Energy usage data as well as sales data estimates must be obtained for an accurate estimate
of savings potential.
Collaborate with other stakeholders to advocate for stronger appliance standards to DOER
and other state entities. Set up informational meetings with state regulators and
Legislators to provide information briefings and outline plans going forward.
Page 7 of 8
2.0 SAVINGS METHODOLOGY FOR CODE COMPLIANCE SUPPORT INITIATIVE The Codes and Standards initiatives proposed by the Massachusetts PAs are relatively new. As a result,
established savings and protocols for evaluation have not previously been developed. There is no existing
framework that currently has been implemented for calculating savings associated with code compliance
support activities2.
This section is divided into four categories. The first section describes the process of estimation for gross
Technical Potential (GTP) savings for the 2016‐2018 plan period; the second describes the estimation for gross
savings for the 2016‐2018 plan period; the third section describes the theory behind PA attribution and net
savings for this same period; and the fourth describes the evaluation process. The methodologies provided
below have been derived from a similar methodology that was created for a similar initiative in National Grid’s
Rhode Island energy efficiency programs. The RI version of the methodology has been reviewed in detail by
NMR Inc. in the report titled “Rhode Island Code Compliance Enhancement Initiative Savings and Attribution
Logic Evaluation”, dated 11/7/2013.
The chart below identifies the breakdown and definitions of gross technical potential and net savings that
consist of various factors like PA incentive program savings, NOMAD, and non‐compliance. The savings
methodology and calculations conducted by the MA PAs were also peer reviewed by NMR Inc. The results of the
calculations below incorporate the changes recommended by NMR Inc.
2 Mass PAs have been in constant contact with the DOE funded PNNL project, the DOE funded BCAP project, PG&E and
MEEA to periodically check if a better and more accurate methodology has been created. To date, the PAs believe that our
methodology is the most accurate representative of the savings potential in MA.
Page 8 of 9
Figure 1: Steps involved in Net Savings Calculation
Following are the definitions explaining the chart above:
Gross Technical Potential: This represents the residential and commercial energy savings (kWh and Therms)
through building simulations described in sections below. The gross technical potential for C/I is the difference
between site observed energy measures and buildings modelled as 100% compliant with 2012 IECC
requirements multiplied by the total square feet of new commercial buildings in MA. The Gross technical
potential for residential is the difference between homes modelled with the User Defined Reference Home
(UDRH) used for the 2014 Residential New Construction (RNC) program and homes modelled as 100% compliant
with 2012 IECC multiplied by the total number of single family and multifamily new construction permits in
MA.3 Both C/I and residential numbers represent savings as a result of measures that didn’t comply with the
code (below or worse than code requirements). Measures that exceed code requirements were not included in
the energy simulations. To the extent buildings adopt efficiency measures beyond code requirements these
would be counted separately in the new construction programs, to the extent those customers participated.
Non‐compliance: Non‐compliance (post evaluation):4 The difference between 100% compliance and the
evaluated compliance at the end of the code cycle. For example, if code compliance at the beginning of the PA
code effort was 80% and after 3 years, the evaluation survey indicated 95%, then non‐compliance is 5%. The
3 Note that for the 2016‐2018 plan cycle the PAs are conservatively excluding savings from retrofit/renovation projects in existing
residential and commercial buildings in deemed savings estimates due to lack of data. However, in the future, it is expected that CCSI will
have some impact on bringing more existing buildings into code compliance and may be claimed after further study.
4 The term compliance here refers to the estimates of the energy savings associated with the different levels of compliance and not just a
count of buildings that complied 100% with every single measure/code requirement.
Page 9 of 10
compliance here is being referred to as estimates of the energy savings associated with the difference and not
as a percentage of items in the code.
Gross Savings: The gross savings is the difference between the compliance level of buildings (in energy terms)
before a PA supported code effort and after a certain numbers of years of influence. In this case, the gross
savings is the GTP minus the non‐compliance estimated after evaluation.
Attribution Factors: There are two prime contributors under attribution factor: NOMAD (described below) and
the impact of the CCSI. The PAs and EEAC Consultants determined that savings through above code actions can
be excluded from attribution factors. This is to avoid any double counting of savings between CCSI and the PA’s
new construction programs, which will claim these above code savings, where appropriate.
Naturally Occurring Market Adoption (NOMAD) factors: These include the natural learning curve of
building professionals and education efforts of other entities like the Dept. of Public Safety, DOER,
Manufacturer associations, ASHRAE, USGBC, AIA, etc. NOTE: The residential studies referenced in
the ‘evaluation section’ below (in the absence of a PA administered code compliance initiative)
indicated that the compliance rate at the end of one code cycle was 76% and the compliance rate at
the beginning of the next cycle was 63%.The difference between these two rates could be attributed
to NOMAD and education through other entities besides PAs.
It should be noted that the use of compliance levels play an important role in the calculation of
savings. In order to develop these levels a methodology must be used to translate findings from on‐
site measurements into an overall average value. The method that was used in the initial studies
was the PNNL (Pacific Northwest National Laboratory) checklist which at the time was the best
available method. In order to be consistent across studies and time this method will continue to be
used, although if a better alternative is determined that method will be pursued.
The PAs propose that the second component of the “attribution factor(s),” which reflects the portion of gross
savings attributable to the PA code Compliance Support component, would be 1 minus the NOMAD factor, and
is listed above and would in the future be determined by a Delphi panel for prospective use. The PAs plan to use
a deemed value for this three year planning cycle (2016‐2018), also described in detail in sections below.
New Construction versus Retrofit Buildings:
This memo highlights the gross and net savings pertaining to new construction buildings only, although this
initiative targets both new construction and retrofit markets. The PAs believe that there may be a higher non‐
compliance rate in retrofits than new construction, though the impacts of energy savings for the retrofit market
not complying with the code, and the PAs ability to affect this, are currently unknown. The PAs are currently
working with EEAC consultants, our contractors, and other national experts to determine how savings from
retrofits could be applied to this methodology. Some of the issues with the savings methodology pertaining to
the retrofit market are as follows:
Page 10 of 11
There is no accurate way of estimating retrofit permits from the ‘total’ permits that are currently
available in public databases.
There is no accurate way of estimating what % of the retrofit permits is applicable for energy
components of the buildings and what that would be in terms of energy savings. The PAs have
incorporated this as a survey question during our energy code trainings and we hope to have better data
on the number of retrofit permits and % attributed to energy components of these permits.
On‐site surveys for retrofit buildings were beyond the scope of the existing code baseline studies. So,
there doesn’t exist any code baseline studies for retrofit buildings at the time of writing this memo. The
energy savings as a result of non‐compliance will have to be based on new construction modeling
results, or in the future, the PAs may include retrofit surveys as part of the upcoming code baseline
study scopes.
The market penetration of the renovation/rehabilitation PA administered program is currently unknown
and will have to be estimated to deduct the program savings from the gross savings.
While the code calls for retrofit projects to bring associated square footage and equipment up to code, it
seems likely that many projects in the retrofit market may not be in full compliance and the overall
compliance rate may be quite different than that for new construction. Further, the CCSI may have a
greater influence on new buildings than on this retrofit market. As a result, not only is the gross
technical potential from retrofit uncertain, but the ultimate CCSI impact is also uncertain at this time.
Future evaluations will attempt to better inform estimates of retrofit savings.
For the purpose of the 2016‐2018 program planning, the PAs will exclude retrofit energy savings and focus only
on new construction. During the course of the next three years, PAs will continue to develop a methodology for
estimating retrofit savings and monitor the influence of the CCSI in the retrofit market.
2.1 EstimatesofGrossTechnicalPotentialAs part of the C&S program planning process, the PAs developed preliminary estimates of the gross technical
potential that could result from the proposed tasks. The numbers included here are estimates based on the
most recent, best available information. The compliance support initiative has two distinct energy savings
estimates, for commercial and residential buildings, respectively.
(a) 2.1.1ResidentialSectorInputs from the Massachusetts User Defined Reference Home (UDRH) used for the 2014 RNC program were
used to develop a building energy model, and simulations were run to compare energy consumption with that of
the same building prototype built to 2012 IECC prescriptive code specifications by different fuel type (propane,
oil, natural gas, electricity). The energy impact was separated into estimates of kWh and Therms for HVAC,
Page 11 of 12
DHW, and Lighting (kWh), and then multiplied by the number of single family and low‐rise multifamily
residential new construction units for Massachusetts as estimated by the 2014 U.S. Census results5.
Residential Buildings Modeling Approach
The analysis used 168 sample homes to represent various configurations (apartments, townhomes, and single‐
family detached homes of various sizes); heating system types (air‐source heat pumps, boilers, and furnaces);
heating fuels (electricity, gas, oil, and propane); and locations (Worcester, Boston, and Pittsfield). Each
characteristic was assigned a weight based on its prevalence in (a) the MA Residential New Construction
Program in 2014, and (b) state‐wide analyses. Each of these homes was modeled using characteristics consistent
with the 2012 IECC residential code, and they were then simulated using REM/Rate, a common home energy
modeling program used by HERS raters. Also simulated were corresponding reference homes for each IECC 2012
home. Reference homes were based on the latest Massachusetts UDRH (user‐defined reference home) used in
the 2014 RNC program, which is in turn based on a baseline study conducted in 2012. Differences in energy
consumption between the UDRH and 2012 IECC homes were compiled by end use and/or fuel type, and final
savings values were weighted according to the home characteristics described above. Final savings results (i.e.,
consumption differences between 2012 IECC and MA UDRH) were then compiled and reported by fuel type, and
state‐ vs. program‐wide distributions.
In Appendix A of this document, modeling results of a single family home, a townhome and a multifamily
dwelling unit are provided.
Approach to Calculation of State‐wide savings for Residential Buildings
The following section was paraphrased from NMR’s peer review memo dated July 17, 2015:
The savings were weighted to reflect the distribution of newly constructed single‐family detached and single‐
family attached homes in the Northeast. The census bureau offers detailed splits on newly constructed single‐
family detached and single‐family attached homes only at the regional level. According to the Census, in 2013
and 2014, 80% of all newly constructed single‐family housing was detached and 20% was attached;6 these are
the weights that were applied in the savings calculation.
Massachusetts‐specific Census Bureau permit data was used to update the growth rate and permit counts for
single‐family new construction. The growth rate of single‐family permits in Massachusetts in 20147 was 3 % but
due to the variability of this growth rate between 2011 and 2014 the PAs, in consultation with the EEAC
Consultants agreed to use a conservative growth rate of 0% for the years 2015 through 2018. The evaluation
team recommends that these numbers be updated annually to reflect actual growth rates and permit counts.
The last step in the single‐family review was to incorporate the penetration rate of single‐family homes from the
RNC program. The penetration rate was included so that program homes could be excluded from the residential
5Source: http://www.census.gov/construction/bps/txt/tb2u2014.txt 6 http://www.census.gov/construction/nrc/pdf/quarterly_starts_completions.pdf 7 http://www.census.gov/construction/bps/txt/tb2u2013.txt and http://www.census.gov/construction/bps/txt/tb2u2014.txt
Page 12 of 13
savings calculations8. The 2013 penetration rate was 29% for single‐family homes; this value is assumed for the
period of 2014 through 2018 as well.
Low‐rise multifamily buildings are typically associated with the residential energy code and therefore are
included in the residential savings calculations; high‐rise multifamily units are typically associated with the
commercial energy code and as a result are captured in the commercial savings calculations.9
Census Bureau data was used to estimate the number of permits and the growth rate for low‐rise multifamily
construction from 2013 to 2017. The growth rate of multifamily permits in Massachusetts from 2013 to 2014
was ‐4.4%. In consultation with the EEAC Consultants the PAs used a growth rate of 0% for 2015 through 2018.
Projecting growth rates is challenging as the number of permits has fluctuated significantly over the last few
years due to the collapse and subsequent rebound of the economy. As is the case with the single‐family data,
the evaluation team recommends that these numbers be updated annually based on actual growth rates and
permit counts.
The Census Bureau does not differentiate between low‐rise and high‐rise multifamily permits. Therefore, the
NMR team estimates that 50% of the multifamily permits are low‐rise multifamily units and belong in the
residential savings calculations. Finally, the 2013 RNC program penetration rate of 30% for multifamily units was
applied to the years of 2014 through 2018. Including the program penetration rate allows multifamily program
units to be subtracted from the savings calculations10.
Summary:
Baseline Model = 2014 RNC program UDRH energy model
Proposed model = A 2,798 sq. ft. single family 2012 IECC fully compliant home in MA, a
2,347 sq. ft. town home and a 1,595 sq. ft. multifamily low‐rise dwelling unit.
Broken down by fuel type and end use measure
Final results capture kWh and Therms. Other fuels like oil and propane were not reported.
Simulations are done using REM/Rate11. Only those measures in the UDRH model that were
below code (non‐compliant) were reported in the energy savings potential
8 Even though only savings from areas that were measured as falling below (non‐compliant) code requirements are counted in the CCSI
and savings from areas above code are counted in the Residential New Construction (RNC) program; savings for the CCSI needed to be
discounted for the penetration of the RNC program. The UDRH used in the modeling of energy consumption utilized information from
the baseline study which only included non‐program homes. Therefore it was not representative of the entire new construction
population of both program and non‐program homes and would have resulted in savings that were too high if not discounted.
9 Low‐rise multifamily captures units in buildings with three stories or less while high‐rise multifamily captures units in buildings with four
stories or more. 10 See Footnote # 8.
11 REM/Rate™ Software is used by organizations which conduct Home Energy Rating Systems (HERS®) Ratings.
Page 13 of 14
Difference between the two is ‘gross technical potential energy savings’ per home.
To determine energy savings for residential homes in MA, we used census data to
determine number of single family and low‐rise multifamily permits in 2013 and multiplied
number of permits by the energy savings per home.
Assumes a three year energy code cycle and one additional year from the end of code cycle,
making it four years total
Annual savings indicate savings per year based on the # of permits/yr
A 0% annual growth rate for single family and multi‐family was assumed across all 4 years.
FORMULA: State‐wide gross technical potential in kWh & Therms = (Difference between
energy consumption of 2014 RNC program UDRH & 2012 IECC 100% compliant home) X
number of residential permits
Figures 2 & 3 below show the electric and gas Gross Technical Savings for residential buildings that are
permitted under the 2012 IECC code. Assuming a code cycle of three years, it is reasonable to assume that it will
take up to four years for homes to fall within one code cycle (some homes permitted under one code cycle may
complete construction at the beginning of another code cycle). “Year 1”, indicated in the table below, is
assumed to be the year when a new code has been in effect for six months (in MA, the 2012 IECC became
mandatory on July 1, 2014, so ‘Year 1’ is 2015). Each year has savings associated to it based on the number of
new buildings constructed under the prevailing code. A cumulative four year total has also been provided in the
table below. To align with the PA three year plan cycle of 2016 to 2018, the PAs will claim savings (see
attribution section below) by combining 2015 and 2016 (Year 1 + year 2) as combined 2016 savings, Year 3 for
2017 and Year 4 for 2018.
Figure 2: Single Family Residential Gross Technical Potential
Figure 3 Low‐Rise Multifamily Residential Gross Technical Potential
Single Family Residential Code
Compliance Energy Savings
Gross Technical Potential Savings
New Construction Residential Single FamilyYear 1 (2015) Year 2
(2016)
Year 3
(2017)
Year 4
(2018) 4‐year total
NC Electric "Annual" Savings (MWh) 8,965 8,429 8,132 7,955 33,481
NC Gas "Annual" Savings (Therms) 1,362,696 1,362,696 1,362,696 1,362,696 5,450,784
MWh/Therms
Low‐rise Multifamily Residential
Code Compliance Energy Savings
Gross Technical Potential Savings
New Construction Residential MultiFamilyYear 1 (2015) Year 2
(2016)
Year 3
(2017)
Year 4
(2018) 4‐year total
NC Electric "Annual" Savings (MWh) 3,350 3,093 2,950 2,864 12,257
NC Gas "Annual" Savings (Therms) 37,643 37,643 37,643 37,643 150,572
MWh/Therms
Page 14 of 15
(b) 2.1.2CommercialSectorSimilar to the residential methodology, the baseline for the commercial building sector was determined as
buildings that meet 100% of the 2012 IECC code, and were then compared to non‐compliant buildings that were
surveyed during the 2012 code baseline study (commercial buildings on average were 80% compliant with the
2006/2009 codes at the time of the study in terms of energy savings). New Buildings Institute12 conducted
building modeling simulations for five building types based on data collected during the 2012 code baseline
study. Energy Use Intensities (EUI) for offices, schools, multifamily, retail and refrigerated warehouses were
created both for 100% compliant conditions and for those when compliance was not met. The EUIs were then
multiplied by the forecasted number of square feet of new construction commercial buildings in MA using the
online Dodge Database. The assumption is that commercial buildings have a long cycle from start to completion.
So the savings associated with one code cycle will be stretched over 5 years until permits are generated under
the next new code. In the table below, Year 1 is assumed as 2015, six months after the new 2012 IECC took
effect.
Commercial Buildings Modeling Approach Converting Field Data to Modeled Results: The first step was to convert the collected field data (from the code baseline study done in 2012) into building energy use characteristics. It is to be noted that a number of building features were listed in the field results as yes/no compliance data, providing no basis on which to determine specific building characteristics for the energy modeling analysis. For air barriers, a number of observations were made in the field about the air barrier, but the only technical information provided was the air leakage rate of the glazing system. For HVAC systems, a great deal of effort was required to convert observations about HVAC compliance into a set of characteristics that could be modeled.
As suggested in the evaluation section of this document, in future baseline studies it would be possible to design a field protocol to collect information on building characteristics that would be focused more specifically on building characteristics that inform energy modeling.
Observations: The results show that on the whole, the energy performance of the building stock significantly exceeds (outperforms) basic code requirements for each building type. The energy improvement over code requirements ranges from 10% to over 20% better than code, depending on building type and assumptions about missing data. This occurs because although individual building components may not meet code requirements, other components of the buildings significantly exceed code requirements, and the energy impact of these features far outweighs the energy impact of non‐complying features.
If the energy associated with non‐compliant characteristics is considered alone (without considering offsetting impacts of over‐compliance), the energy impact of these non‐compliant components ranges from 5% to 8% worse than buildings that just meet code. In a sense, this analysis category represents an opportunity to reduce energy use by improving code compliance of specific building features, with the caveat that for buildings submitted under the performance path in the code, these characteristics may not represent non‐compliance. The table below shows the EUI results through modeling of site observed buildings (only non‐compliant components were extracted), the EUI of buildings that are 100% compliant with 2012 IECC (3rd column) and the technical potential EUI (4th column)
12 New Buildings Institute, Discussion of Draft Results for Massachusetts Energy Code Compliance Impact Analysis October 27, 2014
Page 15 of 16
Building Type
Baseline (site
observed modeled)
2012 IECC EUI (100% compliance)
Technical Potential Savings EUI
Apartment 56.2 50.2 6.0
Office 62.4 55.6 6.8
Retail 59.1 53.6 5.5
Education 45.2 42.5 2.7
Storage 29.6 26.4 3.2
Other 47.1 42.0 5.0
Weighted Avg: 52.0 46.9 5.1 Figure 4: Difference Between Energy Use Intensity (EUI) of Site Observed Modeled Building and Code Compliant Buildings
The NBI analysis provided a very different perspective on the impact of energy code compliance in Massachusetts compared to other studies that have focused specifically on compliance rates. While it is true that a number of buildings demonstrated non‐compliant characteristics, this analysis suggests that the energy impact of these non‐compliant characteristics is significantly offset by a pattern of over‐compliance in other building features. There are patterns of non‐compliance with specific code requirements that represent an opportunity to increase energy savings by focusing on improving compliance rates with these requirements. A detailed memo was provided to the PAs by NBI that can be made available on request.
Growth Rate and Determination of Square Feet of New Construction:
The NMR memo dated July 27, 2015 proposes that the 3‐year average change in Construction Gross Domestic
Product (GDP) for MA from the US Department of Commerce, Bureau of Economic Statistics13 be used as the
new construction growth rate in the commercial savings calculator. The rate of change in new construction
starts as documented by Dodge reports is highly variable from year to year while the MA Construction GDP is a
more stable and replicable estimate of the future commercial new construction growth rate in the state.
The NMR memo also recommends revisions to the estimates of commercial project volume (ksf14) built by year
to reflect projects completed under the IECC 2012 rather than project starts. The revised estimates of project
completions by year assume commercial new construction projects take between one and five years to
complete and completed project ksf are distributed uniformly over that time span.
Based on feedback received from National Grid new construction sales representatives and our new
construction evaluation contractors, typically small commercial projects take from 6 months to 2 years from
start to completion of projects, while larger projects can take up to 3 years from start to completion. The
estimates suggested by NMR appeared to be low based on a construction completion cycle of small and large
13http://www.bea.gov/iTable 14 thousand square feet
Page 16 of 17
projects. Hence, National Grid proposes the following distribution of square feet (net savings in the table 4
below are calculated using this breakdown of square feet). This table assumes that 50% of 2015 buildings will be
completed at the end of 2016 and remaining 50% at the end of 2017. Fifty percent of buildings starting in 2016
will be completed at the end of 2017 and another 50% at the end of 2018.
Figure 5: Table Showing Estimates of Square Feet of Construction within the Code Cycle
Summary:
IECC fully compliant model: Modeled 5 building types that are 2012 IECC fully compliant
buildings in MA. Results are in EUI per building type (office, multifamily, retail, school,
warehouse)
Broken down by fuel type and end use measure
Site observed model: Modeled the same 5 buildings types with building energy features
observed on site. Results are reflected in EUI
Note: only the ‘non‐compliant’ energy savings were accounted for. Measures that
exceeded code were not captured in the model and would be counted where
appropriate in the new construction program outside of CCEI (refer to NBI’s memo for
details, see footnote number 4)
Used Dodge data for sq. ft. of commercial buildings based on 2013 numbers (see
detailed description above)
Assumes 5 years for commercial projects to be completed under a 3 yr code cycle. In the
calculation, for each year, total square footage with a start date in that year is used in
the calculations.
Annual savings indicate savings per year based on sq. ft./yr
A flat 2.6% annual growth rate was assumed across all 4 years (per NMR memo)
FORMULA:
– State‐wide savings in kWh & Therms = (Difference between site observed energy
consumption & 2012 IECC 100% compliant building) X total Sq. ft. of new buildings
The table below indicates the gross technical potential for commercial buildings that are permitted under the 2012 IECC code. Year 1 starts with 2015. As with the residential calculations, to align with the PA three year plan cycle of 2016 to 2018, the PAs will claim savings (see attribution section below) by combining 2015 and 2016 (Year 1 + year 2) as combined 2016 savings, Year 3 for 2017 and Year 4 for 2018. Some savings for this code cycle will be realized in 2019 but has not been listed here.
Assumed percentage of ksf completed under IECC 2012 each year
New construction growth rate 2.6% 2013 2014 2015 2016 2017 2018
ksf of Comm new construction starts‐ 2014 19,756 0% 0% 25% 25%
ksf of Comm new construction starts‐ 2015 20,270 0% 0% 50% 50%
ksf of Comm new construction starts‐ 2016 20,797 0% 0% 50% 50%
ksf of Comm new construction starts‐ 2017 21,338 0% 0% 50%
ksf of Comm new construction starts‐ 2018 21,893 0% 0% 0% 0% 0% 0%
‐ ‐ ‐ 15,074.10 25,472.62 21,067.40 ksf of Comm new construction completed under IECEC 2012
Page 17 of 18
Figure 6: Commercial Buildings Gross Technical Potential
2.2 EstimatesofGrossSavingsThe gross savings is the difference between the compliance level of buildings (in energy terms) before a PA
supported code effort and after a certain numbers of years of influence. In this case, the gross savings is the GTP
minus the non‐compliance estimated after evaluation. Non‐compliance (for the purpose of calculating savings
for PA effort in 2016‐2018) is assumed as 10% for commercial buildings and 17% for residential buildings.
Below, gross residential electric savings are derived in an example calculation (single family only).
Gross Savings
= ∗
2016 Net PA Savings (MWh)
= 8,965MWH ∗. .
.8,429MWH ∗
. .
.
= 4,846 4,556
= 9,402
2017 Net PA Savings (MWh)
= 8,132MWH ∗. .
.
= 4,396
2018 Net PA Savings (MWh)
= 7,955MWH ∗. .
.
= 4,300
Commercial Code Compliance Energy
Savings
Gross Technical Potential Savings
New Construction CommercialYear 1 (2015) Year 2
(2016)
Year 3
(2017)
Year 4
(2018)Year 5 (2019)
NC Electric "Annual" Savings (MWh) ‐ 13,207 22,318 18,459 N/A
NC Gas "Annual" Savings (Therms) ‐ 313,171 529,204 437,684 N/A
MWh/Therms
Page 18 of 19
Where;
Gross Technical Potential (GTP) ‐ is derived as described above
Non‐Compliance = 0.17, this is an estimate, it assumes the residential ending compliance rate (or Target
CCSI Compliance, see below) with CCSI intervention will be 0.83, therefore non‐compliance = 1‐0.83 =
0.17.
Baseline Compliance = 0.63, this is based on the following report, “Massachusetts Electric and Gas
Program Administrators Code Compliance Results for Single‐Family Non‐Program Homes in
Massachusetts, NMR Group, Inc., September, 2, 2014.”
The gross savings listed in Figure 7 below are used in the calculation of net savings indicated in Section 2.3
below.
Gross Savings for Code Compliance Support by PAs Annual Electric and Gas Gross Savings
Electric Gross Savings
SF Residential Savings (MWh)
MF Residential Savings (MWh)
Commercial Savings (MWh)
Elec Yr 1 Svgs (2015 & 2016) 9,402 3,483 7,704
Elec Yr 2 Svgs (2017) 4,396 1,594 13,019
Elec Yr 3 Svgs (2018) 4,300 1,548 10,768
Elec Svgs ‐ 3 Yr Total 18,098 6,625 31,491
Gas Gross Savings
SF Residential Savings (Therms)
MF Residential Savings (Therms)
Commercial Savings (Therms)
Gas Yr 1 Svgs (2015 & 2016) 1,473,185 40,695 182,683
Gas Yr 2 Svgs (2017) 736,592 20,348 308,702
Gas Yr 3 Svgs (2018) 736,592 20,348 255,316
Gas Svgs ‐ 3 Yr Total 2,946,370 81,390 746,701
Figure 7 Electric and Gas Gross Savings for Residential and Commercial Code Compliance Support
2.3AttributiontoPAEffortsandNetSavingsOnce savings are achieved, there is the question of attribution. Obviously, there are different entities involved
in developing, endorsing, and promoting codes and standards; with each contributing to the level of success
Page 19 of 20
associated with the codes and standards in effect. California has the longest standing program and associated
attribution process. California’s process involves a rigorous methodology appropriate for a program responsible
for 9% to 15% of their EE portfolio savings. The California codes and standards program evaluation attribution
approach relies on Delphi panels of industry experts, who estimate both factor weights and scores to determine
the portion of savings that is allocated to the IOUs. The savings occur from the development of requirements
and are affected by compliance rates (among other factors). However, California has not yet tried to claim the
savings associated with compliance support.
Arizona has developed a process where a pre‐negotiated percentage of savings are claimed by the participating
utilities. The Arizona Corporation Commission has agreed to allow utilities to claim 1/3 of the verified energy
savings resulting from codes and standards support in their territories. Utilities still have to demonstrate that
the savings are being realized through an evaluation, measurement and verification (EM&V) study, and
document that they were involved in supporting the codes, but the exact attribution rate is predetermined.
In the absence of existing detailed information to inform future attribution, the PAs will use the following
assumed attribution factors. The numbers below will be deemed and used for the 2016‐2018 planning purpose.
PAs propose that more accurate attribution factors be estimated by a Delphi panel during the course of the next
three years (2016‐2018) for prospective use going forward only.
PA Net Savings through Code Compliance Support: The following formula was used for projecting net savings
from CCSI for the 2016 – 2018 Plan, for each year of the plan.
Gross Savings * Attribution Factor * Annual Ramp Factor.
Where,
Gross savings ‐ is derived as described above
Attribution Factor – The percentage of potential energy savings above the normal compliance level, on
average, at the end of a typical energy code cycle attributable to PA CCSI efforts.
Annual Ramp Factor – Factor used to simulate how quickly the CCSI reaches the target compliance goal
across years. That is, since it takes time for the education efforts of the CCSI to take hold only a portion
of the attributable savings are claimed each year during the initiative and ramped up to 100% over the
entire three year term.
The second term in the above formula, Attribution Factor, adjusts the Gross savings to account for only those
savings that may be attributable to the PAs energy efficiency programs due to the intervention of the CCSI. This
could be thought of as Potential Attributable savings.
The third term, Annual Ramp Factor, accounts for the ramp up of the CCSI efforts across the years. The
Potential Attributable savings are adjusted to account for the impact of the CCSI in a given year. The following
percentages are assumed for the years of implementation.
Page 20 of 21
Year 1 = 20%
Year 2 = 30%
Year 2 = 50%
Year 4 = 100%
Note, PA’s are proposing conservative values for the first two years because final non‐compliance will only be
estimated through EM&V in the 3rd year. This reduces any risk of over claiming early savings and allows for
potential capture of the full savings only in the final year of the 3‐year plan cycle depending on EM&V results.
Below, net residential electric savings are derived in an example calculation (single family only).
For this calculation, the 2012 IECC cycle is assumed to last three years and end in June 2017. Since the 2012
IECC cycle began July 1, 2014 and the CCSI has been running since mid‐2014, savings from 2015, 2016, 2017 and
2018 were incorporated into the calculation for savings during the 2016‐2018 Plan term.
2016 savings were the sum of 20% of 2015 savings (1st year of initiative implementation) and 30% of
2016 savings (2nd year of initiative implementation).
2017 savings were 50% of 2017 savings (3rd year of initiative implementation).
2018 savings were 100% of 2018 savings (4th year of initiative implementation).
Net Savings
= ∗ ∗
2016 Net PA Savings (MWh)
= 4,846MWH ∗ 0.35 ∗ 0.2 4,556MWH ∗ 0.35 ∗ 0.30
= 339 479
= 818
2017 Net PA Savings (MWh)
= 4,396MWH ∗ 0.35 ∗ 0.50
= 769
2018 Net PA Savings (MWh)
= 4,300MWH ∗ 0.35 ∗ 1.0
= 1,505
Where;
Page 21 of 22
Attribution Factor = 0.35. This is calculated using the following formula;
(Target CCSI Compliance‐Natural Ending Compliance)/(Target CCSI Compliance‐Baseline Compliance)
Where;
Target CCSI Compliance = The estimated compliance level at the end of an energy code cycle
after CCSI intervention, in this case estimated at 0.83.
Natural Ending Compliance = The average compliance level at the end of an energy code cycle
with no CCSI intervention, 0.76 based on the following report, “Massachusetts Electric and Gas
Program Administrators Code Compliance Results for Single‐Family Non‐Program Homes in
Massachusetts, NMR Group, Inc., September, 2, 2014.”
Baseline Compliance = represents the percentage of energy savings realized at the beginning of
a new code cycle.
Ramp Factor = As indicated above.
The PAs propose the following savings for 2016 to 2018 program planning allocated between the PAs.
Net Savings for Code Compliance Support by PAs Annual Electric and Gas Savings with each year as a percentage of total Non‐NOMAD Compliance
Yr 1 = 20%, Yr 2 = 35%, Yr 3 = 60%, Yr 4 = 100%
Electric Savings
SF Residential Savings (MWh)
MF Residential Savings (MWh)
Commercial Savings (MWh)
Elec Yr 1 Svgs (2015 & 2016) 818 302 809
Elec Yr 2 Svgs (2017) 769 279 2,278
Elec Yr 3 Svgs (2018) 1,505 542 3,769
Elec Svgs ‐ 3 Yr Total 3,325 1,208 7,598
Gas Savings
SF Residential Savings (Therms)
MF Residential Savings (Therms)
Commercial Savings (Therms)
Gas Yr 1 Svgs (2015 & 2016) 128,904 3,561 19,182
Gas Yr 2 Svgs (2017) 128,904 3,561 54,023
Gas Yr 3 Svgs (2018) 257,807 7,122 89,360
Gas Svgs ‐ 3 Yr Total 515,615 14,243 162,565
Page 22 of 23
Figure 5: Annual Net Electric and Gas for 2016‐2018 Code Compliance Support Initiative
The PAs and EEAC consultants agreed on a 20 year measure life for the code compliance support initiative. This
is based on much of the energy savings from improved compliance reflecting very long lived measures (often the
entire life of the building) and for savings that may be shorter term such as lighting LPD it seems likely customers
will replace lamps with the same or better technology over time. Figure 5 reflects the lifetime savings associated
with this initiative.
Figure 5: Lifetime Net Electric and Gas for 2016‐2018 Code Compliance Support Initiative
Agreement regarding changes to 2016 – 2018 savings
As noted above in the descriptions of the various terms and example calculations there are many variables used
to estimate savings due to the CCSI. Both the PAs and EEAC Consultants agree that the estimation of savings is
a difficult endeavor that has many associated challenges. These challenges include areas that are difficult to
measure, have a long lead time and an even longer measurement span. The PAs and EEAC Consultants have
collaborated to develop what we believe is a reasonable yet conservative methodology for savings estimation.
Some of the main inputs required for this methodology include;
Electric Lifetime Savings
SF Residential
Savings
(MWh)
MF Residential
Savings
(MWh)
Commercial
Savings
(MWh)
Elec Yr 1 Svgs (2015 & 2016) 16,360 6,040 16,180
Elec Yr 2 Svgs (2017) 15,380 5,580 45,560
Elec Yr 3 Svgs (2018) 30,100 10,840 75,380
Elec Svgs ‐ 3 Yr Total 61,840 22,460 137,120
Gas Lifetime Savings
SF Residential
Savings
(Therms)
MF Residential
Savings
(Therms)
Commercial
Savings
(Therms)
Gas Yr 1 Svgs (2015 & 2016) 2,578,080 71,220 383,640
Gas Yr 2 Svgs (2017) 2,578,080 71,220 1,080,460
Gas Yr 3 Svgs (2018) 5,156,140 142,440 1,787,200
Gas Svgs ‐ 3 Yr Total 10,312,300 284,880 3,251,300
Page 23 of 24
Residential single family new construction permits
Residential multi‐family new construction permits
Commercial new construction square footage
Compliance with the various code requirements at the beginning of a code cycle pre CCSI intervention
Compliance with the various code requirement at the end of a code cycle pre CCSI intervention
Compliance with the various code requirements at the beginning of a code cycle post CCSI intervention
Compliance with the various code requirements at the end of a code cycle post CCSI intervention
Modelling of residential buildings based on typical building practice and 100% code compliant
Modelling of commercial buildings based on typical building practice and 100% code compliant
Due to the amount of information required, the degree of difficulty in obtaining it, the time required to acquire it, the various policy rules that affect the factors used in the calculations, the PAs and EEAC Consultants have agreed on what can and cannot change for the savings associated with the CCSI in the 2016‐2018 Plan. All variables assumed in the calculation of projected savings will be held constant with two exceptions. This includes holding constant the deemed attribution factor, which is analogous to a net‐to‐gross factor, and therefore is a consistent approach with all other MassSave savings claims. Future attribution factors will be refined based on EM&V results and a Delphi panel, for use prospectively in the following plan cycle.
The first exception is the compliance level at the end of the 2012 IECC code cycle (also referred to as the “Target CCSI Compliance” in the example calculation above), which will be determined by a study most likely to take place somewhere in the 2017‐2018 timeframe. For projecting savings in the 2016‐2018 Plan, estimates of 0.83 for residential and 0.90 for Commercial were developed. Once a study is completed the new “evaluated” estimates may end up higher or lower and will be substituted into the savings calculation to determine final savings.
The second exception is the “Baseline Compliance”, or the compliance level estimated or measured at the beginning of the 2012 IECC Code cycle. Obviously the goal of the CCSI is to support energy code compliance resulting in an overall improvement in the compliance level. There are a few different factors which can impact the Baseline compliance, including the following two which are worth noting.
1) The level of change between codes (e.g. a large change between codes might lead to a lower Baseline Compliance in the next code, while a small change might lead to a higher Baseline Compliance).
2) The long term effect of the CCSI efforts (e.g. the expectation is that the CCSI will have a positive effect on both the awareness of the energy code, as well as, a carryover effect to the next code cycle).
This evaluation effort is trying to measure the impact of CCSI and therefore needs a baseline or compliance level prior to intervention. Both the PAs and EEAC Consultants feel that the best information possible has been used to estimate the Baseline Compliance (as well as all other factors) as accurately as possible. The Baseline estimate used for the 2016‐2018 Plan is based on a study prior to CCSI intervention and where the level of change between code cycles is approximately consistent over the last few energy code cycles. If a measurement from a different code cycle is used, it may not be an “apples to apples” comparison. Therefore the PAs and EEAC Consultants agree that this variable will only be revised if both parties agree that it would be reasonable to change (e.g. if one party agreed but the other did not, then the variable would not change).
In essence, due to the timing and effort required for the various studies as mentioned above, this means that the 2016 and 2017 planned savings will be unchanged for reporting purposes, but that 2018 savings will be
Page 24 of 25
impacted by the results of the compliance studies completed by the end of 2018. To the extent possible, based on any new information and the results of studies completed prior to the filing of the 2019‐2021 Plan, all factors will be updated for the projections used to estimate savings during the 2019‐2021 time frame.
Page 25 of 26
3.0 EVALUATION Even with all of the nationwide activity occurring in the codes and standards area, PA‐driven initiatives to
compliment, enhance, and build upon them are relatively new. The components, structure, and benefits to the
various stakeholders are still being developed in many jurisdictions. There are only a few programs that are up
and running from which to draw experience. The method for estimating, evaluating, and verifying the savings
associated with these activities remains a work in progress.
The PAs have developed a multi‐faceted approach to evaluation of the codes and standards initiative
incorporating baseline studies, surveys, data/material collection, and proposed Delphi panels. The evaluation
efforts will feed both the attribution process as well as the savings estimations. Since the codes and standards
initiative is a long‐term effort that will require years to conduct and years to realize the benefits, an overall
Evaluation Plan has been developed to help coordinate the effort. The Plan incorporates existing, current, and
future work. Please note that future evaluation plans are based on the best information currently available, but
are subject to change should key stakeholders determine there are alternative evaluation approaches that
would result in a more accurate estimation of the CCSI impacts.
Another important aspect as evaluation of the CCSI moves forward is cost control. Conducting
baseline/compliance studies, energy modelling, collecting various survey data and analysis tends to be time and
labor intensive. As the initiative and evaluation move forward economies of scale and efficiencies will be sought
in order to keep costs within reasonable limits and potentially streamline the efforts. As this is a new initiative
costs will tend to be much higher at the outset but hopefully will be able to be reduced as more information and
trends are better understood.
3.1PreviousEvaluationEffortsEvaluation of the CCSI begins with work previously conducted and incorporated into the current effort. This
includes baseline/compliance studies in both the residential and commercial/industrial sectors. In 2011, two
residential baseline studies were conducted; one looking at homes built at the end of the 2006 IECC cycle15 and
one looking at homes built at the beginning of the 2009 IECC cycle16. A third study17 was conducted using the
results from these two studies which summarized code compliance rates for single family homes at these
specific time periods (end of 2006 IECC and beginning of 2009 IECC cycles).
A compliance study for commercial and industrial buildings was conducted in 201118. Since the building cycle is
much longer in the C&I sector, this study was not developed to look at buildings at the end or beginning of the
15 Massachusetts Mini Baseline Study of Homes built at the End of the 2006 IECC Cycle, Final Report, NMR Group, Inc., KEMA Inc.,
Dorothy Conant, July 17, 2012. 16 Massachusetts 2011 Baseline Study of Single‐family Residential New Construction, NMR Group, Inc., KEMA, Inc., The Cadmus Group,
Inc., Dorothy Conant, August 16, 2012. 17 Massachusetts Electric and Gas Program Administrators Code Compliance Results for Single‐Family Non‐Program Homes in
Massachusetts, NMR Group, Inc., September 2, 2014. 18 Final Report, Project 11, Code Compliance Baseline Study, Massachusetts Energy Efficiency Programs’ Large Commercial & Industrial
Evaluation, DNV KEMA, Energy & Resource Solutions, APPRISE, August 24, 2012.
Page 26 of 27
code cycle. The study looked at 75 recently completed buildings that ranged in size from less than 25,000 ft2 to
greater than 250,000 ft2, covered multiple building types, and various geographic regions around the state.
About two thirds of the buildings were completed under the 2006 IECC and one third under the 2009 IECC. A
second C&I study building on the previously completed C&I study was just completed19 at the end of July 2015.
This C&I study has taken a deeper dive and analysis of the data from the previous study and looked at 50
additional buildings that are in the construction stage. The sample set included 25 buildings under 50,000 ft2
and 25 buildings over 50,000 ft2. By utilizing data from the previous C&I study and this just completed study
compliance levels for the beginning and end of the 2009 IECC cycle were developed. While the earlier C&I study
completed in 2012 did not look at Stretch Code due to the timing of adoption of the Stretch Code and building
completion, this recently completed study did. Of the 50 sites 33 were in Stretch Code communities. Separate
compliance rates were calculated for Stretch Code versus non‐Stretch Code sites and even with the more
stringent Stretch Code requirements the compliance rate for the Stretch Code sites was higher.
Since the initial residential and commercial baseline/compliance studies were conducted, an overall Evaluation
Research Plan, as mentioned above, was developed in August 201420 and amended in May 201521. This Plan was
developed during the planning stages of the CCSI. It covers the short‐term CCSI evaluation efforts in detail while
taking a much higher perspective of longer term evaluation dealing with aspects of the Codes and Standards
Initiative which may be implemented in the future. As such, the Evaluation Plan is considered a living document
that may change as conditions on the implementation front change. In addition to covering short‐term
evaluation methodologies, the Evaluation Plan also describes and recommends an attribution methodology
making use of Delphi panels.
3.2CurrentandOn‐goingEvaluationEffortsMentioned in the high level description within the Plan are baseline/compliance studies that are currently
ongoing for the residential sector. A residential baseline/compliance study scheduled to take place during the
spring/summer/fall of 2015 is currently in the field recruiting homes and conducting on‐site visits. This study
will look at single family homes built at the end of the 2009 IECC cycle, 2009 IECC Stretch Code, and the
beginning of the 2012 IECC cycle. This is the first residential baseline study that includes inspections of stretch
code homes and the data resulting from this study will be used to inform the evaluation of CCSI impacts. Moving
forward, stretch code and non‐stretch code impacts will be calculated separately to account for any differences
in baseline characteristics and CCSI influence.
With respect to the CCSI, the Evaluation Plan details how the various implementation efforts undertaken will be
documented. For the training portion, the evaluation in 2015 includes: (1) tracking the number of trainings by
19 Massachusetts Commercial New Construction Energy Code Compliance Follow‐Up Study, Final Report, DNV‐GL, ERS, APPRISE, July 22,
2015.
20Massachusetts Electric and Gas Program Administrators, Detailed Research Plan – Cross‐Cutting Code Compliance Support Initiative
Evaluation, NMR Group, Inc., August 12, 2014. 21 Massachusetts Electric and Gas Program Administrators, Amended Research Plan for 2015 – Cross‐Cutting Code Compliance Support
Initiative Evaluation, NMR Group, Inc., Cadmus Group, and Tetra Tech, May 5, 2015.
Page 27 of 28
type and location and the number of trainees by category and job titles/descriptions; (2) immediate surveys
after training to assess what the trainees learned and what they expect to apply; (3) follow‐up interviews to see
what the trainees recall and what they have actually applied in practice; (4) process assessments of the
residential and commercial trainings; and (5) assessment of residential and commercial code compliance
methods by different municipalities. For the Circuit Rider portion, evaluation includes such things as: ensuring
that the appropriate data is collected and immediate surveys.
Attendees at every classroom training since September 2014 are asked to complete a short paper survey at the
end of the session assessing the usefulness of the information presented, whether the information was new to
them, and the overall quality of the training. Over 800 residential and over 300 commercial training attendees
had completed these surveys by the summer of 2015. The PAs receive summaries of the immediate survey
responses six times a year: three covering residential trainings and three covering commercial trainings.
Eighty follow‐up in‐depth interviews are currently being conducted with training attendees approximately six
months after the trainings: 60 with residential and 20 with commercial training attendees. The follow‐up
interviews seek to determine how much of the respondents’ recent work has made use of the trainings and how
the trainings are being used in the field. The PAs will receive the findings of the follow‐up interviews by the end
of 2015.
The evaluators are attending three residential and three commercial trainings in 2015 to assess the level of
comprehensiveness and detail presented on the technical subject materials, the structure and pace of training
delivery, the quality of presentations and materials, the time spent on various topics and sufficiency of class
duration, and attendee questions and concerns. The PAs will receive the findings of the training process
assessment by the end of 2015.
The evaluators are also visiting 50 building departments to collect residential data and six building departments
to collect commercial data documenting energy code compliance and verifications by code officials. The PAs will
receive the findings on the information collected on code compliance by different municipalities by the end of
2015.
The evaluators are conducting short telephone surveys of individuals who contact the CCSI with questions
through the telephone or by e‐mail as soon as their cases are closed; the respondents rate the service in terms
of quality, usefulness, and how well issues were resolved. The PAs receive summaries of these survey responses
four times a year.
3.3FutureEvaluationPlanningFuture baseline/compliance studies will look at the end of the 2012 IECC cycle and stretch code for residential
single family and at the end of the 2012 IECC cycle plus stretch code for Commercial and Industrial. In addition,
methods to collect information will be researched that will better align with the energy modelling that will need
to take place in order to determine potential and actual savings. The exact timeframe to have the residential
and commercial studies conducted will depend on when the 2012 IECC cycle ends. The chart below gives an
indication of how the timeline for the various baseline studies might occur across years and energy code cycles.
Page 28 of 29
This chart assumes that the 2012 IECC cycle would end June 2016, and that the 2015 IECC cycle would begin July
1, 2016.
In the longer term, the Evaluation Plan calls for Delphi Panels to be conducted in late 2017 and early 2018 to
determine attribution rates for estimated CCSI savings. This currently proposed methodology is flexible; if a
better alternative is identified, it will be incorporated in the next year or so. The Delphi process involves sending
a first‐round questionnaire to a number of experts, putting together a summary of responses, and then re‐
circulating the questionnaire accompanied by the results from the first round. NMR (the CCSI evaluation
contractor) successfully used a Delphi panel to estimate savings attribution to the Residential New Construction
(RNC) Program in 2013‐2014. Eight of the 14 panelists were building energy efficiency experts with extensive
experience working in Massachusetts; the remaining panelists managed RNC programs in other states or were
2006 IECC 2009 IECC Code Cycle 2012 IECC Code Cycle 2015 IECC Code Cycle Future code cycle
Qtr: 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
Yr:
2013‐2015 Three Year Plan 2016‐2018 Three Year Plan
Legend
‐ This color indicates planning for 3 Year Plan
A: Mini‐Baseline Conducted (residential end of 2006 IECC cycle)
B: Full Baseline Conducted (residential beginning of 2009 IECC cycle)
C: Large C&I Code Compliance study (included buildings built under 2006 & 2009 IECC)
D: Study to determine energy code compliance related to A & B
E: 2014 LCI study ‐ 50 buildings in different stages of construction
F: Residential Baseline study to look at end of 2009 IECC Code cycle, 2009 IECC Stretch Code, and beginning of 2012 IECC cycle
G: 2013 ‐ 2015 Plan Year(s) Report
H: Residential Baseline study to look at end of the 2012 IECC cycle and the beginning of the 2015 IECC cycle
I: Commercial Baseline study to look at buildings built during the 2012 IECC cycle
J: Attribution work ‐ Delphi panel (begin July 2018, finish December 2018 ‐ may need to move up for planning purposes)
K: File 2016 ‐ 2018 Plan Year(s) Report
L Residential Baseline study to look at end of 2015 IECC cycle, beginning of Future cycle?
M: Commercial Baseline study to look at buildings built during the 2015 IECC cycle
20152010 2011 2012 2013 2014 2016 2017 2018 2019 2020
CCSI Implementation CCSI Implementation
Page 29 of 30
nationally known efficiency program evaluators. The RNC study used a diverse panel that achieved some
convergence once they examined each other’s reasoning.
By the time of the proposed Delphi panels, the CCSI portion of the Codes and Standards Initiative will have been
in place for close to 4 years and other portions such as Appliance Standards Advocacy, may come into play.
There will be a substantial amount of information to provide to the Delphi Panel for consideration. These inputs
include documentation of all the trainings, marketing efforts, Circuit Rider activities, and activities from other
components of the CCSI. In addition to this implementation data, the findings from the 2015 evaluation work
discussed earlier—all the immediate surveys, training process assessment, code compliance documentation,
and the follow‐up interview report—as well as additional evaluation findings from later years will be
summarized for the Delphi panelists. The panelists will also receive an overview of activities by other groups,
which may influence code compliance. From the baseline studies described earlier, the panelists would also
receive information on various energy efficiency characteristics of newly constructed homes and buildings and
pre‐ and post‐CCSI compliance values for both the residential and C&I sectors.
Using all of this information, the Delphi panels would retrospectively identify the changes in construction
practices that are attributable to the CCSI and other influences, possibly including the RNC and C&I New
Construction Program and other PA programs. The Delphi panels would estimate the counterfactual values—
that is, how homes and buildings would have been built in the absence of the programs considered. The
evaluation team would then model energy use for the baseline homes and buildings as they were built (“as‐
built”), and then model assuming the counterfactual practices to estimate energy use; in other respects the
building models would be the same. The difference between the as‐built and counterfactual models would be
the estimated net energy savings. To simplify the process, above‐code or above‐prevailing practice savings
could be attributed to the RNC and C&I New Construction Programs, and savings from getting buildings closer to
code or to the current prevailing practice could be attributed to the CCSI.
After the modeled retrospective savings estimates are complete, the evaluators will ask the Delphi panels to
develop prospective net‐to‐gross estimates. Materials to help in their deliberation will include: the retrospective
savings estimates; description of expected program activities; description of expected code changes; and
discussion of other things that could influence building practices, such as naturally occurring market adoption
(NOMAD), LEED building, energy prices, economic conditions, and climate change. All of the aforementioned
evaluation activities will look at stretch code and non‐stretch code impacts separately to account for the fact
that the influence of the CCSI may be different depending on which building code is in effect in various
municipalities.
Because National Grid is also pursuing a very similar initiative in Rhode Island and undergoing EM&V activities
there, every attempt will be made to leverage information, as appropriate, from Rhode Island and reap
economies to Massachusetts to most efficiently direct EM&V resources.
Page 30 of 31
4.0 APPLIANCE STANDARDS ADVOCACY Projecting savings from advocacy of appliance standards also is a unique challenge. In order to determine the
savings from any particular standard, the specific technology or product will need to be clearly defined and
thoroughly researched. These research activities could take significant time. Even if the specific technology and
associated savings were known, there could be zero actual savings during the 2016 – 2018 Three Year Plan due
to the time lag between the beginning of research and advocacy efforts and the installation of qualified products
and therefore realized savings. The PAs have collaborated with NEEP (Northeast Energy Efficiency Partnership)
and ASAP (Appliance Standards Awareness Project) to determine a potential appliance list that PAs can offer
advocacy and technical support for appliance adoption. The savings estimates in the table below are based on
engineering estimates of the savings for each appliance unit between the current standard practice and a
stringency level identified by the Kaboom II report22 produced by ASAP, and also included in ASAP’s proposal for
MA’s model bill. Market penetration is based on retail sales of efficient units, and estimated to remain steady
from the second year onward. It is estimated that half of the units sold in the first year that the appliance
standard is in effect will meet the new energy efficiency requirements. It is important to remember that the
annual energy savings will continue to grow because: a) once an appliance is installed, it saves energy for the
lifetime of that unit, and b) new units are installed each year. Savings would conclude when the appliances’
lifetime concludes, or a new federal standard is adopted, pre‐empting the standard the PAs helped to develop
and get adopted.
The table below is for illustrative purpose only and indicates that if MA adopts all the appliances listed below in
the model bill, then in 2018, there are 128GWh of electric savings and 30 BBTU of gas savings available to the
state for adoption at the state level. Assuming PAs contribute to advocacy of these appliances and create
technical specifications for additional appliances not currently in this list, with a 10% attribution to PA effort in
appliance standards development and advocacy, PAs can potentially claim 12,800 MWh of electric savings and
30,000 Therms in 2018.
At this point we propose that deeming of attribution rates and gross technical potential savings from standards
be developed at the time that specific appliance standard efforts are planned, based on the specific standards
and activities expected. Therefore, at this time the PAs have not included any planned savings for the 2016‐2018
Plan cycle.
22Appliance Standards Awareness Program, Kaboom II Report
Page 31 of 32
Page 32 of 33
APPENDIX
Figure 8: Modeling Results of Savings for a Single Family Home (Going from UDRH to Code)
Savings by Primary Heating Fuel
Pro
pan
e W
eig
hte
d S
avin
gs
Oil
Wei
gh
ted
Sav
ing
s
Nat
ura
l G
as W
eig
hte
d S
avin
gs
Ele
ctri
c W
eig
hte
d S
avin
gs
To
tal
Wei
gh
ted
Sav
ing
s
2012 IECC/MA UDRHState Wide Housing Distribution
% of Market Served 7.3% 11.3% 79.3% 2.3% 100.0%
Weighted Average Home CFA 2,798 2,798 2,798 2,798 2,798
Heating (kWh) 313.14 315.11 313.14 4786.70 414.02
Heating (MBtu) 32.30 32.26 32.30 0.00 31.57
Cooling (kWh) 140.8 140.9 140.8 124.3 140.4
DHW (kWh) 0.0 0.0 0.0 -1.5 0.0
DHW (MBtu) -0.1 -0.1 -0.1 0.0 -0.1
Lights and Appliances (kWh) 0.0 0.0 0.0 0.0 0.0
Total Electric Savings (kWh) 454.0 456.0 454.0 4,909.5 554.4
Total Fossil Fuel Savings (MBtu) 32.2 32.2 32.2 0.0 31.5
2012 IECC/MA UDRHState Wide Housing Distribution
Page 33 of 34
Figure 9: Modeling Results of Savings for a Townhouse Home (Going from UDRH to Code)
Figure 10: Modeling Results of Savings for a Low‐rise Multifamily Home (Going from UDRH to Code)
Savings by Primary Heating Fuel
Pro
pan
e W
eig
hte
d S
avin
gs
Oil
Wei
gh
ted
Sav
ing
s
Nat
ura
l G
as W
eig
hte
d S
avin
gs
Ele
ctri
c W
eig
hte
d S
avin
gs
To
tal
Wei
gh
ted
Sav
ing
s
2012 IECC/MA UDRHState Wide Housing Distribution
% of Market Served 0.8% 12.0% 80.7% 6.5% 100.0%
Weighted Average Home CFA 2,347 2,347 2,347 2,347 2,347
Heating (kWh) 69.04 70.56 69.04 4174.28 337.43
Heating (MBtu) 5.23 5.21 5.23 0.00 4.89
Cooling (kWh) 21.2 21.2 21.2 76.3 24.8
DHW (kWh) 0.0 0.0 0.0 -1.2 -0.1
DHW (MBtu) 0.0 0.0 0.0 0.0 0.0
Lights and Appliances (kWh) 0.0 0.0 0.0 0.0 0.0
Total Electric Savings (kWh) 90.2 91.8 90.2 4,249.5 362.1
Total Fossil Fuel Savings (MBtu) 5.2 5.2 5.2 0.0 4.9
2012 IECC/MA UDRHState Wide Housing Distribution
Savings by Primary Heating Fuel
Pro
pa
ne
We
igh
ted
Sa
vin
gs
Oil
We
igh
ted
Sa
vin
gs
Na
tura
l Ga
s W
eig
hte
d S
av
ing
s
Ele
ctr
ic W
eig
hte
d S
av
ing
s
To
tal W
eig
hte
d S
av
ing
s
2012 IECC/MA UDRHState Wide Housing Distribution
% of Market Served 0.8% 12.0% 80.7% 6.5% 100.0%
Weighted Average Home CFA 1,595 1,595 1,595 1,595 1,595
Heating (kWh) 27.00 27.74 27.00 1004.40 90.94
Heating (MBtu) 1.51 1.90 1.51 0.00 1.46
Cooling (kWh) 22.4 22.4 22.4 131.2 29.5
DHW (kWh) 0.0 0.0 0.0 163.1 10.7
DHW (MBtu) 0.1 0.0 0.1 0.0 0.0
Lights and Appliances (kWh) 0.0 0.0 0.0 0.0 0.0
Total Electric Savings (kWh) 49.4 50.1 49.4 1,298.6 131.1
Total Fossil Fuel Savings (MBtu) 1.6 1.9 1.6 0.0 1.5
2012 IECC/MA UDRHState Wide Housing Distribution
Page 34 of 35
Figure 11: Weighting Factors Used in Energy Models for Residential Buildings
Weighting Factors:
Characteristic
Primary Heating
Fuel Type
Electric 6.5% 2.3%
Natural Gas 80.7% 79.3%
Propane 12.0% 11.3%
Oil 0.8% 7.3%
Primary HVAC
System Type
Air‐Source Heat Pump (Ducted) 8.2% 5.0%
7.0% 7.0%
Forced Air (Ducted) w/ CAC 84.8% 88.0%
Boiler w/ Radiant Heat w/ no C
Value
Weighting
Percentage
(Program‐wide)
Weighting
Percentage
(State‐wide)
Weather Location
Boston 50.2% 50.2%
Worcester 38.8% 38.8%
Pittsfield 11.0% 11.0%